What is the type of my problem?

Hi
I’m a beginner in the field of GNN and have some questions.

I want to develop a model which recieves many graph swith some features for nodes and edges as the input and one feature of the nodes as the output in the training phase.

My goal is to train a GNN model that recieve a graph and predict that feature in the nodes.

That feature is a boolean value that determines a nodes feature is True or False.

Is my problem a node classification or graph classification?

I myself think is node classification and read that part in tutorial but couldn’t find any distingushable statemnts for defining inputs and outputs. ( also for prediction )

Is that tutorial for unsuprrvised classification?

What do you mean “one feature of the nodes”? Do you have one feature per graph or one feature per node in a graph?

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.